MediSeven.jl
A robust Julia package for medical named entity recognition (NER) using spaCy models. MediSeven.jl provides state-of-the-art extraction of medical entities like drugs, dosages, frequencies, and routes from clinical text.
Features
- 🏥 Medical NER: Extract DRUG, DOSAGE, FREQUENCY, DURATION, ROUTE, FORM, and STRENGTH entities
- 🔄 Automatic Fallbacks: Robust model loading with multiple fallback options
- ⚡ Batch Processing: Efficient processing of multiple texts
- 🤖 spaCy-Powered: Built on spaCy and spaCy models
- 📊 Production Ready: Comprehensive error handling and logging
Installation
using Pkg
Pkg.add("MediSeven")
Or from the Julia REPL:
] add MediSeven
Quick Start
using MediSeven
# Load model (automatic fallback if primary model fails)
model = load_model()
# Process single text
doc = model("Take 500mg paracetamol every 6 hours for pain")
println("Entities found: $(length(doc.ents))")
# Process batch
texts = [
"Administer 40mg lisinopril daily",
"Apply 0.1% tacrolimus ointment BID"
]
docs = model(texts)
API Reference
Model Loading
# Default loading with fallbacks
model = load_model()
# Specify model and batch size
model = load_model(model_name="kormilitzin/en_core_med7_lg", batch_size=16)
# Disable fallbacks (strict mode)
model = load_model(model_name="my_model", use_fallback=false)
Text Processing
# Single text
doc = process!(text, model)
# or
doc = model(text)
# Batch processing
docs = process!(texts, model)
# or
docs = model(texts)
Entities
for entity in doc.ents
println("$(entity.label): '$(entity.text)' at $(entity.start):$(entity.stop)")
end
Supported Entity Types
- DRUG: Medication names (aspirin, metformin)
- DOSAGE: Amounts (500mg, 2 tablets)
- FREQUENCY: How often (daily, BID, every 6 hours)
- DURATION: How long (for 7 days, until symptoms resolve)
- ROUTE: Administration route (oral, IV, topical)
- FORM: Drug form (tablet, ointment, injection)
- STRENGTH: Concentration (0.1%, 20mg/ml)
Fallback Models
MediSeven.jl automatically tries multiple spaCy models in order of preference:
kormilitzin/en_core_med7_lg (primary) en_core_med7_lg en_core_web_sm en_core_web_md en_core_web_lg
Development
git clone https://github.com/liyakhathshaik/MediSeven.jl.git
cd MediSeven.jl
julia --project=. -e "using Pkg; Pkg.instantiate()"
julia --project=. -e "using Pkg; Pkg.test()"
Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit changes (
git commit -m 'Add amazing feature') - Push to branch (
git push origin feature/amazing-feature) - Open a Pull Request
License & Attribution
This project is licensed under the MIT License - see the LICENSE file for details.
This package is a Julia port of the original Med7 Python package by Andrey Kormilitzin et al., which is licensed under the Apache License 2.0.
Original Med7 Python Package Attribution
Copyright 2020 Andrey Kormilitzin
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
Citation
@software{mediseven_jl,
title = {MediSeven.jl: Medical Named Entity Recognition for Julia},
author = {Liyakhath Shaik},
year = {2025},
url = {https://github.com/liyakhathshaik/MediSeven.jl},
note = {Julia port of Med7 Python package by Andrey Kormilitzin et al.}
}